Psych Stats #1

Front 1

Two types of Stats

Back 1

Inferential, and Descriptive

Front 2

Variable

Back 2

Characteristic that can have different values

Front 3

Value

Back 3

Possible numbers/categories that a score can have.

Front 4

Score

Back 4

Particular persons value on a a variable

Front 5

Numeric (Quantitative) Level of Measurement

Back 5

Two types: Equal-Interval and Rank-Order

Front 6

Equal-Interval

Back 6

Ratio scale (ex. # of siblings or GPA)

Front 7

Rank Order

Back 7

Ordinal variables. Class rank or winner in a race. Order carries magnetude

Front 8

Nominal (Categorical) Variable

Back 8

Gender; Nominal; No numerical value

Front 9

Discrete Variables

Back 9

Have specific values (Age in years). Cannot have values in between values (No fractions or decimals. All whole numbers/integers)

Front 10

Continuous Variables

Back 10

Have infinite number of values between two values

Front 11

Moduality

Back 11

The number of modes a data set has

Front 12

Unimodal Distributions

Back 12

A data set with only one mode

Front 13

Bimodal Distributions

Back 13

A data set with 2+ modes. Are particularly disruptive when measuring central tendency.

Front 14

Rectangular Distributions

Back 14

More than one mode, but all pretty evenly dispersed

Front 15

Heterogeneous Samples

Back 15

Sample of nonsimilar popular observations. (Example: age of grad school and undergrad students).

Front 16

Symmetrical Distribution

Back 16

Have tails on both sides. One mode and is in the middle.

Front 17

Positive Skew

Back 17

Tail on the right. Data skewed towards left. Median is lower than the mean.

Front 18

Negative Skew

Back 18

Tail on the right. Data majority on the left. Median higher than the mean,

Front 19

Normal Distribution

Back 19

Symmetrical and Unimodal frequency distribution

Front 20

Kurtotic (Pointy)

Back 20

Most data over one or two values. Slim and tall curve

Front 21

Kurtotic (Flat)

Back 21

Data almost evenly split among data. Very round.

Front 22

Floor Effect

Back 22

Drastic positive skew; doesn't accurately measure extent of research.

Front 23

Ceiling Effect

Back 23

Drastic Negative skew; doesn't accurately measure extent of research.

Front 24

Measures of Central Tendency

Back 24

Most "Typical"/Common score. Score best represents the whole distribution. Mean, Median and Mode

Front 25

Mode

Back 25

Most frequently occurring number in a distribution. Measure of CT for Nominal variables. Main problem: Can have more than one mode.

Front 26

Median

Back 26

Midpoint of Distribution. Separates low 1/2 and high 1/2. Not sensitive to extreme scores and outliers.

Front 27

Mean

Back 27

Sum of all scores divided by n. M=Ex/n. Most stable-- all #s included in calculations. Not affected by addition/deletion of scores. Used in many statistical procedures.

Front 28

Varience

Back 28

SD2=E(x-m)/n

Front 29

Z Score

Back 29

the number of standard deviations a score is above or below the mean of the scores in the distribution. Z=x-m/SD

Front 30

Raw Score

Back 30

A regular score before it has been converted to a Z score. Even on different variables can be converted into Z scores and directly compared. X=(Z)(SD)+m

Front 31

Correlation

Back 31

Can be thought of as descriptive statistics for the relationship between two variables. Describes relationship between two equal-interval numeric variables.

Front 32

Scatter Diagram

Back 32

Graph that shows the pattern of relationships of two variables.

Front 33

Linear Correlation

Back 33

Relationship between two variables that shows up on a scatter diagram as dots approximating a straight line.

Front 34

Curvilinear Correlation

Back 34

Any association between two variables other than a linear correlation. Relationship between two variables that shows up on a scatter diagram as following a systematic pattern that is not a straight line.

Front 35

No Correlation

Back 35

No systematic relationship between two variables.

Front 36

Positive Correlation

Back 36

High scores go with High scores; Low scores go with low scores; medium---> medium scores; When graphed, line is up and to the right.

Front 37

Negative Correlation

Back 37

High scores with low scores; When graphed, line is down to the right.

Front 38

Strength of Correlation

Back 38

How close dots fall to a straight line determines the strength of the correlation

Front 39

Best Fit Line

Back 39

f(x)=ax+b

Front 40

Correlation Coefficient

Back 40

Pearsons (r) tells general trend of relationship between two variables. + means correlation is positive. - means correlation is negative; Ranges from 0-1. 1, -1 mean variables are perfectly correlated. 0 means no correlation.

Front 41

Correlation Coefficient Formula

Back 41

r=E(ZxZy)/n

Front 42

Possible Directions of Correlation Coefficient

Back 42

X--> Y -- Less sleep means more stress. Y-->X -- more stress means less sleep. Z--> X+Y -- working longer hours causes both stress and less sleep.

Front 43

Statistical significance

Back 43

A correlation is statistically significant if it is unlikely that you could have gotten a correlation as big as you did if in fact there was no relationship between variables. If probability is less than some small degree of probability (5%), the correlation is considered statistically significant.

Front 44

Predictor Variable

Back 44

X Variable. Independent Variable.

Front 45

Criterion Variable

Back 45

Y Variable. Dependent Variable.